Modern litigation often involves two separate, extra-legal features: (1) contingency fee arrangements with the plaintiff-side attorney, and (2) a "battle of the experts" where the outcome of the case rests on conflicting expert witness testimony. We construct a model to illustrate that when these features occur simultaneously, in contexts where plaintiff-side attorneys bear the cost of hiring experts upfront, defendants will spend systematically more than plaintiffs on hiring expert witnesses. Our model is motivated by hand-collected data from a decade of securities class action litigation, where we show that experts polarize into almost entirely plaintiff-side versus defense-side experts, that defense-side experts are more "prestigious" in terms of educational and academic affiliations, and that defense-side experts are at least 36% more expensive in per-hour terms. We argue that expert asymmetry could be closed through optimal contracting in certain individual claims, but it is likely a necessary feature of all class actions, such as toxic torts or consumer protection, in an "adversarial" litigation system. This defense-side advantage therefore provides a newly identified rationale for "inquisitorial" litigation systems, including increased use of court-appointed experts under Federal Rule of Evidence 706 and state law analogues.
Because judges exercise discretion in how they handle and decide cases, heterogeneity across judges can affect case outcomes and, thus, preferences among litigants for particular judges. However, selection obscures the causal mechanisms that drive these preferences. We overcome this challenge by studying the introduction of random case assignment in a venue (the U.S. District Court for the Western District of Texas) that previously experienced a high degree of case concentration before one judge (Alan Albright), whom litigants could select with virtual certainty. To assess Albright's importance to patent enforcers, we examine how case filing patterns changed following the adoption of random case allocation and show that case filings in the Western District of Texas decreased significantly at both the intensive and extensive margins. Moreover, to shed light on why litigants prefer Judge Albright, we compare case management metrics and motions practice across randomly assigned cases and show that cases assigned to Albright received relatively early trial dates and generated fewer motions to stay pending parallel administrative invalidity proceedings and fewer motions to invalidate patents on subject matter eligibility grounds.
Does judicial ideology affect the process of agenda-setting in the High Court of Australia? Applications for "special leave" to appeal are the primary method through which the High Court manages its caseload, analogous to certiorari at the U.S. Supreme Court. Acceptance rates are low (10%-15%) and the consequences of rejection for litigants are high, since the Court rules as the final court of appeal in all areas of Australian law, including constitutional matters. To process special leave applications, the Court selects small, variably sized, panels of justices to adjudicate, deciding cases primarily via oral argument or summary disposition. However, the specific criteria for panel selection are nowhere explicitly documented, and by practice, the Chief Justice chooses the size and composition of the special leave panels. We argue this creates clear potential for ideological bias to impact case outcomes. Applying novel methods, we demonstrate that, despite substantial variability in the ideological composition of special leave panels, there is very limited evidence that panels are selected strategically to limit the success of applications in particular areas of law, and that there is no evidence that ideological composition has any impact on the success of special leave applications. Our findings contribute to a growing literature on the politics of judicial agenda setting outside the United States.
Judges seek to maximize their own utility, like everyone else. Their goals include job security and promotion. Federal judges lack the electoral incentives that often drive state judges, but they could audition for promotion. I test whether they audition for promotion in their criminal sentencing. Using criminal sentences imposed in the federal courts from 2006 to 2023, I find that federal judges who are eventually nominated to a higher court hand down longer sentences than their other "contender" peers, though this effect is moderated by the judicial district. This effect is a mixture of judicial behavior and a selection effect-judges are promoted for a variety of other reasons, many of which may be correlated with criminal sentencing. While there is no across-the-board evidence of judicial auditioning during vacancies, there is evidence of judicial auditioning for some judges under some presidents. This behavior, though, ends at nomination: an event-study design finds that the announcement of a nomination has no substantive effect on the nominee's sentencing. These results show that, despite the different structure of the federal judiciary from state courts, strategic rationality on the part of judges may also shape federal criminal sentencing.
Studies of class action settlements have been notoriously difficult because of the logistical burden of assembling datasets from district court orders. This is especially true for non-securities settlements. Moreover, even when they have been undertaken, they have been of unknown representativeness. Yet, since the enactment of the Class Action Fairness Act of 2005 ("CAFA"), defendants have been required to send a notice to the United States Department of Justice every time they settled a class action, lest the settlement not be binding. These notices therefore have the potential to overcome many of the challenges and limitations of past catalogs of settlements. This paper is the first to assess the efficacy of the Department's CAFA notices to study class action settlements. Using Freedom of Information Act requests, I obtained a log the Department keeps of CAFA notices. Although the log contains thousands of entries and many more settlements per year than any prior dataset, I find that the log is unpredictably incomplete. This is especially apparent for securities class actions, but appears to be true of non-securities class actions as well. As such, the log will not improve the representativeness of class action settlement studies. But the log will mitigate the logistical burden of creating datasets and therefore should improve the frequency of settlement studies.
Crime doesn't pay. Or does it? We study the role of expectations regarding sanctions and the likelihood of detection on whether people obey the law. We examine how expectations influence whether people obey the law and conduct simulations of various enforcement counterfactuals. We find the average assessment of the likelihood of detection is reasonably accurate, but those who (mistakenly) believe the probability is lower than it is are much more likely to break the law. Further, expectations with regard to the likely consequences of getting caught are also heterogeneous. In our simulations, perceived fines have little impact on willingness to break the law, but a higher perceived likelihood of apprehension has an appreciable impact. Because marginal respondents are pivotal in the rate of law-breaking, debiasing expectations among the whole population has little impact.
What do you want to do with your property when you die? This paper presents the results of a survey in which we asked this question to a nationally representative sample of 9000 American adults. We gathered data on respondents' families and then asked them how they would like to divide their property among those they leave behind. We find that people are more ambivalent about gifts to spouses and more generous to nonmarital partners than the law of intestacy generally assumes. People also give less to parents and more to siblings, extended relatives, and friends than expected and much more to stepchildren than expected. We find some significant differences by race, class, and gender, with women, African Americans, and people of less income and education preferring to give less to their spouses. Our findings improve upon empirical studies of probated wills by providing an unbiased sample, by including data on demographic characteristics, and by observing the preferences of people in unconventional families. We suggest that although intestacy law is often said to implement majoritarian preferences, it may be more consistent with a mix of majoritarianism, paternalism, and administrative efficiency.
Legal scholars and courts have long viewed disclaimers of affiliation as an ineffective tool for dispelling trademark confusion. We revisit this debate with new evidence from two randomized experiments. Consistent with earlier research, we find that disclaimers alone do not dispel consumer confusion. Our experiments show, however, that simply requiring consumers to register their recognition of a disclaimer can reduce confusion to levels that most courts would consider negligible. We tested two disclaimer acknowledgment tasks in an online shopping context: a burdensome task in which participants were required to retype the text of the disclaimer before they could continue with the study and a much simpler task in which participants were required to click a box affirming that they read and understood the disclaimer in order to continue. The tasks were similarly effective in lowering the probability of confusion to levels below the threshold that would typically trigger a finding of likelihood of confusion.
This article presents the findings of a quantitative study on sentencing practices in Brazil, focusing on the presence of numerical patterns and "penal clustering" in judicial decisions. Drawing on a dataset of criminal sentences from S & atilde;o Paulo-the country's most populous and active judiciary-the research statistically investigates whether Brazilian judges display preferences for certain sentence lengths, such as even numbers, "round numbers," and multiples of 6 months. The study also examines the role of numerical fractions in shaping sentence variability. Despite Brazil's civil law tradition and the absence of formal sentencing guidelines, the results reveal a significant degree of penal clustering, with judges consistently relying on a narrow set of sentencing values. These patterns suggest the influence of cognitive heuristics, particularly "anchoring effects," in sentencing decisions. While the use of numerical fractions may enhance consistency and predictability, it may also constrain individualization-raising concerns about mechanical decision-making and reduced responsiveness to case-specific circumstances. The article discusses the implications of these findings for judicial training, institutional reform, and public policy. It argues for a more data-informed and psychologically aware approach to sentencing, capable of balancing the dual demands of consistency and individualization within both common law and civil law traditions.
While the U.S. healthcare system typically imposes significant out-of-pocket costs on those with insurance, the Affordable Care Act (ACA) requires full coverage of certain preventive health services. Still, one in four eligible Americans remains unscreened for breast, colorectal, and cervical cancer. We hypothesize that a lack of awareness about the ACA's requirement contributes to this gap. Additionally, we investigate whether prior experiences with medical debt deter individuals from seeking even free care, a potential spillover effect of broader healthcare cost burdens. We conducted an online survey experiment with 3354 insured U.S. adults aged 30-74 with moderate household incomes ($30 k-$99 k). We determined whether each one qualified for a free cancer screen based on age and sex, and found that n = 1406 had not received at least one recommended cancer screening. Those participants were randomly assigned to one of four conditions in a 2 x 2 factorial design. In the "free care disclosure" (FCD) treatment, half were informed that ACA-required screenings are fully covered without copays or deductibles. In the "medical debt salience" (MDS) treatment, half were asked about their medical debt history before assessing screening intentions. In addition to measures of intention, the primary outcome was behavioral-whether participants requested a link to take a step toward screening, a proxy for screening behavior, which was not observed directly. As hypothesized, FCD increased screening-related behavior by 5 percentage points (p = 0.031), with effects varying by cancer type. Consistently, 45% of respondents indicated that "costs or coverage" was a common reason for not having gotten screening, and large majorities agreed that "going to the doctor or hospital can be dangerous financially," "the American healthcare system is full of tricks and traps," and "in America, healthcare is never really free" (80%, 65%, and 91% agreeing, respectively). MDS had no direct effect on behavior, but contrary to the hypothesis, individuals with medical debt were more likely to seek screening than those without (p = 0.011). These findings suggest that a lack of awareness about the law providing free preventive care is a significant barrier to screening. Proactive communication by clinicians or public health officials could save lives.
The law often asks judges to determine whether a violation is clearly established or whether legal language clearly or unambiguously supports a position. It demands that litigants not raise arguments or claims that would be easily dismissed. And for their part legal scholars have relied on the distinction between easy and hard cases as they delineate the concept of law itself. Yet, few have empirically studied assessments of legal difficulty. Here, we present the results of two studies that provide insight into the reliability of such assessments from laypeople and actual judges. For both populations, our key finding was this: assessments of case difficulty are highly predictive of the vote breakdown in a case; that is, the larger the size of a majority coalition, the lower the average difficulty rating registered by the members of that coalition (and, conversely, the smaller the size of a minority coalition, the higher the rating registered by its members). These results suggest that, while assessments of case difficulty are tied to case outcomes, they are a complex signal. Experiencing the feeling that a case is unusually difficult is not necessarily an indication that you are dealing with a 50/50 case, as is commonly assumed. Instead, it may be an indication that your position on the merits would be decisively rejected by the vast majority of others confronting the same question.
Many states bar employers from requiring or expecting employees to keep pay secret, and from retaliating against employees who discuss pay. In a nationally representative survey of 2369 US adults in 2021, people living in states with state laws barring pay secrecy reported more supportive attitudes regarding information-sharing and less concern about offending co-workers by asking about salary, but little to no difference in information-sharing behavior, beliefs or accuracy regarding the laws. Men were consistently more supportive of information-sharing than women, but the gender differential did not differ between groups of states. We randomized participants in each set of states to learn the law of their jurisdiction. In states with pay secrecy bans, participants who learned about the law reported significantly greater comfort asking others and intentions to ask others about compensation, compared to those who were not given information about the law.
The American Bar Association has declared a “well-being crisis” among lawyers, but the empirical basis for this claim has been contested in recent years. This study systematically compares two high-quality, nationally representative surveys—the National Health Interview Survey (NHIS) and the National Survey on Drug Use and Health (NSDUH)—to measure the prevalence of mental illness and alcohol misuse among lawyers. In both surveys, lawyers report elevated rates of alcohol misuse compared with the general public and similarly educated peers. The NHIS finds that lawyers experience psychological distress at rates lower than the general public and similar to, or moderately higher than, similarly educated peers. In the NSDUH, by contrast, more than 40% of lawyers report moderate or serious psychological distress in the past year. This rate is significantly higher than those reported by the general public, by similarly educated peers, and the rate found in the NHIS. While we cannot fully explain all of the differences between the two national surveys, we resolve some of these differences by studying sensitivity to instrument validation and calibration and closely aligning the measurements used in both surveys. To assess the remaining differences, we identify several advantages of the NSDUH, including the privacy of data-gathering methods, additional clinically validated mental illness measures, and results that are more consistent with other national surveys. The persistent divergences between the NHIS and NSDUH underscore the challenges of measuring mental illness and the importance of continued work on survey implementation, validation, analysis, and interpretation.
Pretrial risk assessment tools are used in jurisdictions across the country to assess the likelihood of "pretrial failure," the event where defendants either fail to appear for court or reoffend. Judicial officers, in turn, use these assessments to determine whether to release or detain defendants during trial. While algorithmic risk assessment tools were designed to predict pretrial failure with greater accuracy relative to judges, there is still concern that both risk assessment recommendations and pretrial decisions are biased against minority groups. In this paper, we develop methods to investigate the association between risk factors and pretrial failure, while simultaneously estimating misclassification rates of pretrial risk assessments and of judicial decisions as a function of defendant race. This approach adds to a growing literature that makes use of outcome misclassification methods to answer questions about fairness in pretrial decision-making. We give a detailed simulation study for our proposed methodology and apply these methods to data from the Virginia Department of Criminal Justice Services. We estimate that the VPRAI algorithm has near-perfect specificity, but its sensitivity differs by defendant race. Judicial decisions also display evidence of bias; we estimate wrongful detention rates of 39.7% and 51.4% among white and Black defendants, respectively.
Automated legal decision-making is often perceived as less fair than its human counterpart. This human-AI fairness gap poses practical challenges for implementing automated systems in the public sector. Drawing on experimental data from 4250 participants in three public decision-making scenarios, this study examines how different reasoning models influence the perceived fairness of automated and human decision-making. The results show that providing reasons enhances the perceived fairness of decision-making, regardless of whether decisions are made by humans or machines. Moreover, sufficiently individualized reasoning models have a stronger positive impact on the perceived fairness of automated decisions than on the perceived fairness of human decisions. This largely mitigates the human-AI fairness gap. The results thus suggest that well-designed reasons can improve the acceptability of automated governance.
I estimate the relationship between increased creditor rights and legal expenditures of debtor corporations by evaluating a securitization law in India allowing secured creditors to seize collateral. The law reduced spending on legal proceedings used by firms to avoid foreclosure, because debt-related litigation decreased. Firms most affected by the law—with high proportions of tangible assets creditors could seize as collateral—significantly decreased spending on lawyers after the legislation. However, some debtor firms affected by the legislation filed for bankruptcy, either because they failed to successfully renegotiate debt with their multiple creditors or because they took advantage of the automatic stay to prevent creditors from seizing their assets. These firms would have incurred legal costs related to the bankruptcy process. Consistent with this intuition, the post-reform decrease in legal costs is smaller for firms that filed for bankruptcy after the legislation or were already closer to insolvency before the reforms. The results indicate that legal transaction costs associated with foreclosure avoidance are sizable and can be eliminated by strengthening contract enforcement.
A patent system is a central tool in innovation policy. The prospect of monopolistic pricing supposedly encourages firms to innovate. However, there is scant empirical evidence supporting the existence of higher markups for patent-protected products. Using an original dataset that links consumer products to the patents that protect them, we study the impact of patent protection on product prices. Exploiting exogenous variations in patent status, we find that a loss of patent protection leads to an 8%-10% drop in product prices. The price drop is larger for more important patents and is more pronounced in more competitive product markets.
Is settlement of civil litigation good or bad for the legal system? Should we have more or less of it? Should settlement have a private or public purpose? Legal scholars and lawyers have considered these questions for decades, but very little is known about the views of ordinary people. This Article reports results from a nationally representative survey of over 1000 U.S. adults about what respondents thought settlement should look like. Respondents indicated, for example, the extent to which they agreed or disagreed with statements like these: “A settlement between two parties is nobody's business but their own.” “Settling parties are more interested in money than justice.” Our findings convey nuanced lay intuitions about civil settlement, including mixed views about secrecy and settlements, and a tendency to want settling defendants to admit fault and apologize. We also discover that a striking number of lay observers prefer private resolution through contract to oversight by judges. Indeed, a considerable majority thought that at least 75% of disputes would be resolved by settlement in an ideal world, with 100% being the most common choice. The view that more cases should settle was both striking and complicated, highlighting the tension between private resolution of disputes and the public function of litigation and courts.
Is settlement of civil litigation good or bad for the legal system? Should we have more or less of it? Should settlement have a private or public purpose? Legal scholars and lawyers have considered these questions for decades, but very little is known about the views of ordinary people. This Article reports results from a nationally representative survey of over 1000 U.S. adults about what respondents thought settlement should look like. Respondents indicated, for example, the extent to which they agreed or disagreed with statements like these: “A settlement between two parties is nobody's business but their own.” “Settling parties are more interested in money than justice.” Our findings convey nuanced lay intuitions about civil settlement, including mixed views about secrecy and settlements, and a tendency to want settling defendants to admit fault and apologize. We also discover that a striking number of lay observers prefer private resolution through contract to oversight by judges. Indeed, a considerable majority thought that at least 75% of disputes would be resolved by settlement in an ideal world, with 100% being the most common choice. The view that more cases should settle was both striking and complicated, highlighting the tension between private resolution of disputes and the public function of litigation and courts.
